2006 Achievements

التفاصيل

الزيارات: 2880

FIRST PLACE, Egyptian Engineering Day “EED” 2006 IT-Computer Science Challenge

Title: Visual Interpretation of Hand Gestures

Team Members: Ahmed Tarek Ahmed, Hany Mohamed Shafik, Khaled Hany Abd-El-Mageed, Mohamed Faroq Mahdy, Shady Hosny Yussif, and Shady Mustafa Naguib

Supervisor: Prof. Aliaa Abdel-Haleim Abdel-Razik Youssif

Teaching Assistant Assigned: Mr. Amr S. Ghoneim

Abstract:

The project is a computer vision system that recognizes hand gestures in real-time. Current user interfaces are unsuited to harness the full power of computers. Mobile devices like cell phones and technologies such as virtual reality demand a richer set of interaction modalities to overcome situational constraints and to fully leverage human expressiveness. Hand gesture recognition lets humans use their most versatile instrument – their hands – in more natural and effective ways than currently possible. Gesture recognition will allow future human computer interfaces to be more intuitive and user-friendly than traditional interfaces. Gesture recognition with computer vision is non-invasive and more flexible. Yet, it faces difficulties due to the hand’s complexity, lighting conditions, background artifacts, and user differences.

Our goal is to contribute to the available interface modalities and to widen the human-computer interface channel. Leveraging more of our expressiveness and our physical abilities offers new and advantageous ways to communicate with machines.

(Figure Above) The ‘Visual Interpretation of Hand Gestures’ team members in the EED 2006 event, taking a photo with Dr. Ahmed Darweesh, the Minister of state for Administrative Development.

(Figure Above) The ‘Visual Interpretation of Hand Gestures’ team members, with their teaching assistant.

(Figure Above) Al-Ahram Egyptian newspaper, for Thursday January 25, 2007 (Issue # 43879, 35^th page) an article about the proposed Visual Interpretation of Hand Gestures project.

FIFTH PLACE, Made-In-Egypt “MIE” 2006 Undergraduates “Graduation Projects” Challenge

Title: Vehicle Recognition

Team Members: Amira Mohammed Awwad, Amira Mohammed Mohsen, Eman Ahmed Ebrahim, Esraa Ali Hassan, Khadiga Hassan Mahmoud, Nahla Ahmed Ali

Supervisor: Prof. Aliaa Abdel-Haleim Abdel-Razik Youssif

Teaching Assistant Assigned: Mr. Amr S. Ghoneim

Abstract:

Vehicle recognition is the area in computer vision that is concerned with extracting information about vehicles from http://www.fcih.net/main/images or videos; it includes many different categories such as vehicle detection, license plate detection (LPD), license plate recognition (LPR), vehicle’s color recognition and others. The proposed system aims in detecting the vehicle, recognizing its make-and-model (MMR), and detecting its license plate if possible from a single image/video. Vehicle detection is a fairly explored problem, but in contrast the MMR is an unexplored one as it needs a huge database that stores all possible vehicles makes and models to compare the input image with. The proposed system mainly can be divided into four phases;

- Vehicle detection from a static image which aims to detect all vehicles that appear in the image,

- Vehicle detection from videos which aims again to detect the presence of all vehicles, but from videos using the motion detection,

- Make-and-model recognition subsystem takes the segmented image (the one with a detected vehicle) and recognizes its make and model, and

- Finally, the license plate detection subsystem detects the license plate from the segmented vehicle’s image if it possible.

Overall, the system detects the vehicle by extracting the input image’s features by SIFT (Scale Invariant Feature Transform), then clustering the features by k-means clustering, then classify it by a trained SVM (Support Vector Machine) classifier, then localize the vehicle using neighborhood suppression technique with good detection results that is about 81% of tested samples, the output segmented vehicle is recognized by the make and model using 2D template matching technique that searches on a predefined database of models with results 85%, and the LPD uses line extraction algorithm, then rectangular region detection, followed by texture classifier to detect the license plate with about 80% correct results of the test samples. The proposed vehicle recognition system will provide valuable situational information for law enforcement units in a variety of civil infrastructures.

(Figure Above) The ‘Vehicle Recognition’ team members in the EED 2006 event, taking a photo with Dr. Ahmed Darweesh, the Minister of state for Administrative Development.

(Figure Above) The ‘Vehicle Recognition’ team members, with their supervisor.

(Figure Above) Al-Ahram Egyptian newspaper for Thursday February 1, 2007 (Issue # 43886, 32^nd page), an article about the proposed Vehicle Recognition project.